Journal of Applied Psychology © 2012 American Psychological Association

2014, Vol. 99, No. 3, 390 – 403 0021-9010/14/$12.00 http://dx.doi.org/10.1037/a0030264

Leading Virtual Teams: Hierarchical Leadership, Structural Supports, and

Shared Team Leadership
Julia E. Hoch and Steve W. J. Kozlowski
Michigan State University

Using a field sample of 101 virtual teams, this research empirically evaluates the impact of traditional
hierarchical leadership, structural supports, and shared team leadership on team performance. Building
on Bell and Kozlowski’s (2002) work, we expected structural supports and shared team leadership to be
more, and hierarchical leadership to be less, strongly related to team performance when teams were more
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

virtual in nature. As predicted, results from moderation analyses indicated that the extent to which teams
This document is copyrighted by the American Psychological Association or one of its allied publishers.

were more virtual attenuated relations between hierarchical leadership and team performance but
strengthened relations for structural supports and team performance. However, shared team leadership
was significantly related to team performance regardless of the degree of virtuality. Results are discussed
in terms of needed research extensions for understanding leadership processes in virtual teams and
practical implications for leading virtual teams.

Keywords: team virtuality, virtual team leadership, structural supports, shared team leadership, team
performance

Virtual teams work together over time and distance via elec- Most research has focused on the advantages and disadvan-
tronic media to combine effort and achieve common goals (Bell & tages of virtual teams. Relative to face-to-face teams, benefits
Kozlowski, 2002). Although surveys indicate that fewer than 50% attributed to the use of virtual teams include the ability to
of companies used virtual teams in 2000, by 2008 over 65% stated compose a team of experts flung across space and time, in-
that their reliance on virtual teams would “mushroom” in the creases in staffing flexibility to meet market demands, and cost
future. Moreover, among companies with over 10,000 employees, savings from reduced travel (Kirkman, Gibson, & Kim, 2012;
the use of virtual teams was projected to be 80% (i4cp, 2006, Kirkman & Malthieu, 2005; Stanko & Gibson, 2009). Disad-
2008). Concurrent with this growth in the use of virtual teams, the vantages include lower levels of team cohesion, work satisfac-
literature on virtual teams has been increasing (Cheshin, Rafaeli, & tion, trust, cooperative behavior, social control, and commit-
Bos, 2011; Hill, Bartol, Tesluk, & Langa, 2009; Majchrzak, Mal- ment to team goals; all factors that can negatively impact team
hotra, Stamps, & Lipnack, 2004; Martins & Shalley, 2011; performance.
Mesmer-Magnus, DeChurch, Jimenez-Rodriguez, Wildman, & In light of these concerns, it is surprising that relatively
Shuffler, 2011; Peters & Karren, 2009; Sarker, Anjuja, Sarker, & limited research attention has been directed toward virtual team
Kirkeby, 2011; Shin, 2004). leadership (Gibson & Gibbs, 2006; Kirkman et al., 2012; Mar-
tins, Gilson, & Maynard, 2004; O’Leary & Mortensen, 2010;
Siebdraht, Hoegl, & Ernst, 2009). Team leadership is regarded
Editor’s Note. Eduardo Salas served as the action editor for this article.—
S.W.J.K.
as a key mechanism for minimizing motivation and coordina-
tion losses and maintaining team effectiveness when they are
virtual (Bell & Kozlowski, 2002; Malhotra, Majchrzak, &
This article was published Online First December 3, 2012. Rosen, 2007; Martins et al., 2004; Zigurs, 2003). However, one
Julia E. Hoch, School of Human Resources and Labor Relations, Mich- particular concern is that traditional hierarchical leadership
igan State University; Steve W. J. Kozlowski, Department of Psychology,
processes are expected to be disadvantaged in virtual teams
Michigan State University.
The first author would like to acknowledge the German Research Foun-
because of the lack of face-to-face contact. Thus, some scholars
dation (Grant No. 1412/6-1, U. Konradt, PI) for funding that, in part, have suggested that hierarchical leadership processes may need
provided support during her doctoral dissertation research and Dr. Konradt to be supplemented in virtual teams as a way to augment team
for serving as Chair of her doctoral thesis committee. Nonetheless, any effectiveness (Avolio, Kahai, & Dodge, 2000; Bell & Kozlow-
opinions, findings, and conclusions or recommendations expressed are ski, 2002). The purpose of this research is to investigate the
those of the authors and do not necessarily reflect the views of the DFG. impact of team leadership on team performance in teams that
Correspondence concerning this article should be addressed to Julia E. span degrees of virtuality. Although this perspective has been
Hoch, who is now at California State University, Northridge, College of
proposed in the theoretical literature, it has not been examined
Business and Economics, Department of Management, Juniper Hall
JH4216, 18111 Nordhoff Street, Northridge, CA 91330, or to Steve W. J. empirically. In particular, we examine the extent to which
Kozlowski, Michigan State University, Department of Psychology, 309 structural supports and shared team leadership supplement hi-
Psychology Building, East Lansing, MI 48824. E-mail: julia.hoch@csun erarchical leadership and the extent to which these relationships
.edu or je.hoch1@gmail.com or stevekoz@msu.edu are moderated by the degree of virtuality.

390
 LEADING VIRTUAL TEAMS 391

Theoretical Development Structural Supports

• Reward Systems
Leadership in Virtual Teams • Communication and
Information
There is consensus among scholars that virtual teams are more
difficult to lead than face-to-face teams (Bell & Kozlowski, 2002; Hierarchical Leadership Team Performance
Duarte & Snyder, 2001; Gibson & Cohen, 2003; Hinds & Kiesler, • Transformational Leadership
2002; Lipnack & Stamps, 2000). As a consequence of the lack of • Leader-Member Exchange
• Career Mentoring
face-to-face contact and geographical dispersion, as well as the
(often) asynchronous nature of communication, it is more difficult Team
for team leaders to perform traditional hierarchical leadership Shared Team Leadership Virtuality
behaviors such as motivating members and managing team dy- • Cognitive Team Learning • Geographic
namics (Avolio et al., 2000; Bell & Kozlowski, 2002; Purvanova • Affective Team Support Distribution
• Behavioral Member-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

& Bono, 2009). It has been argued that leader influence can be • Electronic Communication
Member Exchange
This document is copyrighted by the American Psychological Association or one of its allied publishers.

extended by having leadership augmented by new media (Avolio • Cultural

Background
& Kahai, 2003; Avolio et al., 2000) and that team leaders simply
have to learn how to use and apply those media properly. Findings
Figure 1. Effects of structural supports, hierarchical leadership, and
from empirical research show that getting virtual teams to function shared team leadership in predicting team performance, moderated by team
equivalently to face-to-face teams requires virtual team leaders to virtuality.
invest much more time and effort (Purvanova & Bono, 2009),
although showing more initiative, trying harder, and investing
more time and energy might not always be feasible. ized as distinct higher-order factors or construct composites, rather
Some scholars suggest that leadership functions should be sup- than unitary constructs. This allows each of the inputs to be concep-
plemented by providing structural supports (Bell & Kozlowki, tualized as a composite of established constructs. For example, hier-
2002; Hinds & Kiesler, 2002; Kahai, Sosik, & Avolio, 2003). For archical leadership is represented by transformational leadership,
example, structuring rewards to provide incentives for perfor- leader–member exchange, and supervisory mentoring. Each of these
mance should result in higher motivation. Another suggested ap- constructs, as core aspects of hierarchical leadership, is supported by
proach is to supplement leadership by distributing leadership to a body of theory and empirical research with established measures.
team members (Bell & Kozlowski, 2002). Sharing leadership with Using established constructs and measures of hierarchical leadership
team members is based on the premise that leadership should not as input factors allows us to clearly assess the potential supplementary
be the sole responsibility of a hierarchical leader, but should be influence provided by structural supports and shared leadership. The
collectively exercised by empowering and developing individual same conceptual and measurement approach using established con-
team members (Kirkman, Rosen, Tesluk, & Gibson, 2004). structs and measures is applied to structural supports and shared
Although this view of leadership challenges in virtual teams has leadership.
consensus in the literature, it has not been subjected to empirical
verification. With respect to improving team performance, it is
Team Virtuality
important to understand the extent to which the influence of
hierarchical leadership is attenuated (or not) as team virtuality With the growth and evolution of virtual teams during the past
increases. Moreover, if the influence of hierarchical leadership is decade, researchers have focused on the conceptualization and
diminished as is suspected, then the extent to which it can be measurement of team virtuality (e.g., Bell & Kozlowski, 2002;
supplemented by structural supports and shared team leadership Hinds, Liu, & Lyon, 2011; Kirkman & Malthieu, 2005). In early
(and, potentially, other supplements) becomes a critical target for research, virtuality was treated as distinctly categorical; research-
theory and research extensions. ers applied a simple dichotomous characterization of virtual and
To examine these issues, our conceptual model treats hierarchi- face-to-face teams. More recently, however, scholars have asserted
cal leadership, structural supports, and shared team leadership as that this simple characterization glosses over a variety of nuanced
inputs to team performance. The model is illustrated in Figure 1. dimensions that underlie a range of differences in the degree of
The basic premise of our approach is that supplementing hierar- virtuality (Gibson & Gibbs, 2006; Irwin & McClelland, 2003;
chical leadership with shared leadership and structural supports Kirkman et al., 2012; MacCallum, Zhang, Preacher, & Rucker,
will be more relevant when teams are more virtual in nature. Thus, 2002; Mesmer-Magnus et al., 2011). Whereas early conceptual-
the degree of team virtuality is predicted to moderate the relation- izations focused exclusively on geographic distribution, subse-
ships between hierarchical leadership, structural supports, and quent conceptualizations added electronic communication and
shared team leadership with team performance. noted differences between the use of asynchronous and synchro-
There are two notable aspects of the model. First, it is focused on nous communications (e.g., Bell & Kozlowski, 2002). Empirical
the contribution of these input factors to team performance. The research, accordingly, refers to both the facets of geographic distri-
model does not focus on mediating processes at this stage of the bution (e.g., O’Leary & Cummings, 2007; O’Leary & Mortensen,
research. The primary reason for this focused approach is to enable a 2010) as well as the relative amount of e-communication media usage
clear evaluation of the moderating effects of virtuality on the contri- (Griffith, Sawyer, & Neale, 2003; Kirkman et al., 2004; Mesmer-
butions of hierarchical leadership, structural supports, and shared Magnus et al., 2011) as indicative of “team virtuality.” This is now
leadership to team performance. Second, the inputs are conceptual- the established approach to conceptualizing virtuality.
 392 HOCH AND KOZLOWSKI

However, virtual teams increasingly span national boundaries are likely facilitated by cues that are more difficult to transmit,
and differences in cultural background are becoming more impor- detect, and interpret in a virtual work context.
tant to consider as an aspect of virtuality (Hinds et al., 2011; Second, LMX also contributes to positive organizational out-
Staples & Zhao, 2006; Tsui, Nifadkar, & Ou, 2007). Indeed, Hinds comes (Gerstner & Day, 1997; Graen & Uhl-Bien, 1995). LMX is
et al. (2011) criticized the lack of inclusion of national and cultural concerned with the nature and the quality of the dyadic relation-
differences in conceptualizations of virtuality. As “organizations ship between the team leader and each member. It describes the
are increasingly compelled to establish a presence in multiple nature of the leader–member relationship and, as such, is primarily
countries as a means of reducing labor costs, capturing specialized developed through face-to-face contact (Gerstner & Day, 1997),
expertise, and understanding emerging markets . . . they often although it can be maintained via forms of electronic communi-
create conditions in which workers must collaborate across na- cation such as e-mail and video-conferencing. LMX provides an
tional boundaries” (Hinds et al., 2011, p. 136). Accordingly, re- alternative mechanism for leader influence (J. M. Howell & Hall-
searchers need to put the global back into “global work” by Merenda, 1999) since interpersonal relationships, once developed,
considering cultural differences. might be less adversely affected by the lack of ongoing face-to-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Research is increasingly considering cultural differences as an face contact in virtual teams, but may also be difficult to develop
This document is copyrighted by the American Psychological Association or one of its allied publishers.

important component of virtuality in globally dispersed teams (e.g., where the leader has little to no face-to-face contact with team
Chen, Kirkman, Kim, Farh, & Tangirala, 2010; Gibson & Gibbs, members.
2006; Tsui et al., 2007). Based on this evolving view of virtuality, our Third, Hamilton and Scandura (2003) highlighted e-mentoring
conceptualization comprises geographic distribution (e.g., O’Leary & as an important leadership function for managing virtual teams,
Cummings, 2007), relative amount of e-communication media usage since it is not restricted to face-to-face contact. Moreover, due to
(e.g., Kirkman et al., 2004), and cultural diversity (e.g., Gibson & virtual interaction, demographic “cues” (e.g., age or gender) are
Gibbs, 2006; Hinds et al., 2011; Tsui et al., 2007) as an addition to the less salient and less likely to influence protégé selection. Accord-
established components of team virtuality. ingly, decisions about who to mentor will be more likely based on
performance criteria. Mentoring further aids in the development of
strong personal relationships that help strengthen leader influence
The Role of Hierarchical Leadership in Virtual Teams
on the team member (Ostroff & Kozlowski, 1993). By increasing
Hierarchical leadership reflects formally designated leadership interaction among leaders and members, it can counteract the
(Ensley, Hmieleski, & Pearce, 2006; Morgeson, DeRue, & Karam, negative effects of limited face-to-face contact in virtual teams
2010; Yukl, 2010). Two well-established leadership theories rel- (Hamilton & Scandura, 2003).
evant to hierarchical leadership that have been widely supported in
the empirical literature are transformational leadership and leader– Hypothesis 1: The positive relationship between hierarchical
member exchange (LMX). Both transformational leadership leadership (transformational leadership, LMX, and mentoring)
(Fuller, Patterson, Hester, & Stringer, 1996; Judge & Piccolo, and team performance decreases as team virtuality increases.
2004; Lowe, Kroeck, & Sivasubramaniam, 1996) and LMX (e.g.,
Gerstner & Day, 1997; Graen & Uhl-Bien, 1995) are strong
The Role of Structural Supports in Virtual Teams
predictors of individual and team performance. Moreover, trans-
formational leadership and LMX are the most prevalent ap- Given that hierarchical leadership is assumed to be more diffi-
proaches used in research on virtual teams (e.g., Avolio et al., cult in virtual teams, it is then important to understand how
2000; Hambley, O’Neill, & Kline, 2007; J. M. Howell & Hall- hierarchical leadership can be supplemented when teams are more
Merenda, 1999; J. M. Howell, Neufeld, & Avolio, 2005). virtual in nature. Structural supports represent a form of indirect
Although it has received less attention, we posit that supervisory influence, where influence on the motivation and behavior of team
career mentoring (e.g., Kram, 1985) is an important leadership members takes place via structural attributes (Bell & Kozlowski,
technique in virtual teams. Supervisory career mentoring is related 2002; Wunderer, 2002). Structural supports draw from the leader-
to career outcomes such as salary level, promotion rate, and job ship substitutes approach (Kerr, 1977; Kerr & Jermier, 1978),
satisfaction, as well as to objective and subjective performance which asserts that aspects of the organization and task structure
(Allen, Eby, Poteet, Lentz, & Lima, 2004; Chao, Walz, & Gardner, can compensate, enhance, or neutralize the effects of leadership on
1992; Scandura & Ragins, 1993; Whitely, Dougherty, & Dreher, employee behavior. While originally proposed as a moderating
1991). Transformational leadership, LMX, and supervisory career variable (J. P. Howell & Dorfman, 1981, 1986), empirical and
mentoring are the three primary constructs that comprise hierar- meta-analytic studies have found strong support for main relation-
chical leadership in the model. ships of structural and compensating variables with team outcomes
First, transformational leadership (e.g., Bass, 1985, 1998) has (Podsakoff, MacKenzie, & Bommer, 1996). Structural factors
been found to enhance performance in a wide range of organiza- have been suggested by Bell and Kozlowski (2002) as a supple-
tional settings (Fuller et al., 1996; Judge & Piccolo, 2004; Lowe et ment for virtual team leadership, which is consistent with structural
al., 1996). Transformational leader behaviors are aimed at inspir- functions listed by other scholars who refer to managing information,
ing follower motivation and stimulating them to stretch their resources, and material rewards (e.g., Fleishman et al., 1991).
capabilities and to go beyond typical performance (Judge & Pic- In virtual teams, the stability and reduction of ambiguity pro-
colo, 2004). However, these forms of leader behavior have also vided by structural supports may compensate for the turbulence
been posited to have weaker relations for virtual teams (Hambley and unpredictability that characterizes virtual teamwork (Zaccaro
et al., 2007; J. M. Howell & Hall-Merenda, 1999; J. M. Howell et & Bader, 2003; Zigurs, 2003). Bell and Kozlowski (2002) argued
al., 2005). Interpretations of leader behavior as transformational that because of the geographic dispersion of virtual teams, an
 LEADING VIRTUAL TEAMS 393

important function of leadership is to create structures and routines (Pearce & Conger, 2003). Thus, sharing leadership functions with
that substitute for direct leadership influence and regulate team team members provides a mechanism to supplement hierarchical
behavior. Consistent with research that suggests structural supports leadership in virtual teams (Bell & Kozlowski, 2002; Pearce, Yoo,
have direct relationships with outcomes that supplement hierarchi- & Alavi, 2004; Tyran, Tyran, & Shepherd, 2003).
cal leadership (Podsakoff et al., 1996), our model conceptualizes Scholars have argued that shared leadership is a more appropri-
them as having a direct relationship rather than a moderating one. ate form of team leadership than hierarchical leadership repre-
Virtual team members usually work on virtual teams in addition sented by the solo leader (Brown & Gioia, 2002; Day et al., 2004;
to their line function and research has highlighted the importance Yukl, 2010). Reasons for this include the notion that team member
of rewarding virtual team members for both aspects. Geographical communication is less formal and less hierarchically based, and,
dispersion can result in a lack of motivation to focus on virtual therefore, team members can more easily overcome communica-
team responsibilities, makes monitoring of virtual team members tion difficulties (Bell & Kozlowki, 2002; Pearce et al., 2004). In
difficult, and also creates higher levels of anonymity (Kiesler & addition, work processes in virtual teams are characterized as
Cummings, 2002; Wiesenfeld, Raghuram, & Garud, 1999). Fur- cognitively loaded, highly interdependent, yet autonomous. Com-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

ther, reward systems need to be fair, such that individual employ- plex teamwork requires the use of self-managing teams (Bell &
This document is copyrighted by the American Psychological Association or one of its allied publishers.

ees perceive they are being rewarded according to their inputs Kozlowski, 2002; Pearce, 2004; Pearce & Manz, 2005). Team
(e.g., effort, time, performance, etc.) on their virtual team work self-management and empowerment, in this context, has been
(Colquitt, 2004; Dulebohn & Martocchio, 1998; Schminke, Cro- shown to enhance virtual team performance in a sample 35 sales
panzano, & Rupp, 2002). Being rewarded in a fair and transparent and service virtual teams in a high-technology organization (Kirk-
way for the work performed on the virtual team will lead employ- man et al., 2004).
ees to put more efforts toward virtual teamwork. There is no “one best way” to measure shared leadership. The
Second, a major component of structural supports is the com- concept is in its infancy (Avolio, Jung, Murry, & Sivasbramaniam,
munication and information management systems used for virtual 1996; Carson, Tesluk, & Marrone, 2007; Mayo, Meindl, & Pastor,
teams. Building and managing communication and information 2003; Mehra, Smith, Dixon, & Robertson, 2006; Pearce & Conger,
management systems that facilitate connectivity, remove percep- 2003), and, thus, a challenge facing researchers is determining how
tions of distance, and facilitate the organization and accessibility of to measure shared team leadership. One primary approach has
information can reduce feelings of lack of trust, anonymity, de- simply treated shared team leadership as analogous to hierarchical
individuation, and perceptions of low social control. In addition, leadership, but conceptualized at the team level of analysis (e.g.,
virtual teamwork is typically white-collar, knowledge based, in- Avolio et al., 1996; Bowers & Seashore, 1966; Pearce & Sims,
tellectual, and interdependent. The management of communication 2002). This approach assesses shared leadership as collective
and information is central to cognitive tasks (Clampitt & Downs, concept in the form of traditional leadership behaviors (e.g., trans-
2004; Faraj & Sproull, 2000). Thus, a key aspect of performance formational leadership) that are performed by team members.
in virtual teams is managing the “triangle” of factors: shared Typically, a traditional leadership measure—like transformational
knowledge (in changing and flexible organization structure), via leadership—is referenced to the team as a collective (reference
electronic communication systems, and with experts as primary shift model; Chan, 1998; e.g., “Our team engages in behaviors that
collaborators (Griffith et al., 2003; Kanawattanachai & Yoo, 2007; help create a team vision”) to comprise shared team leadership.
Malhotra & Majchrzak, 2004). As a form of structural support, However, consistent with other researchers (Carson et al., 2007;
managing communication and information flow (Fleishman et al., Mayo et al., 2003; Mehra et al., 2006), we do not conceptualize
1991) include information infrastructure and quality of informa- shared team leadership as parallel with hierarchical leadership.
tion received, as well as the transparency and adequacy of Team members do not need to necessarily perform the same kind
communication and information management. Communication of leadership behaviors as their supervisors (Künzle et al., 2010;
and information management are posited to influence virtual Morgeson et al., 2010) in order to engage in shared leadership.
team performance. We expect that team virtuality moderates the Rather, shared leadership can be conceptualized as the extent to
relationship between structural supports and team performance. which team members behave in ways to prompt the team processes
Specifically, that underlie team performance. Team process researchers have
distinguished cognitive, affective-motivational, and behavioral
Hypothesis 2: The positive relationship between structural functions as keys to team effectiveness (Kozlowski & Bell, 2003;
supports (reward systems; communication and information) Kozlowski & Ilgen, 2006). Team leader effectiveness, as outlined
and team performance increases as team virtuality increases. in functional leadership (McGrath, 1962), is based on leaders
addressing the cognitive, affective, and behavioral functioning of
The Role of Shared Team Leadership in Virtual their teams (Zaccaro, Rittman, & Marks, 2001). These leadership
functions can be performed through informal leadership mecha-
Teams
nisms (Morgeson et al., 2010) such as shared team leadership.
Shared team leadership describes a mutual influence process, In capturing shared leadership in virtual teams, affective-
characterized by collaborative decision-making and shared respon- motivational functions can be represented in terms of perceived
sibility, whereby team members lead each other toward the team support, which is related to building trust and team cohesion
achievement of goals (Day, Gronn, & Salas, 2004; Pearce & (Kasper-Fuehrer & Ashkanasy, 2001) and may compensate for
Conger, 2003). Shared team leadership is presumed to create specific gaps resulting from the lack of face-to-face meetings in
stronger bonds among the team members; facilitate trust, cohesion, virtual teams, that is, lack of trust, and higher levels of anonymity
and commitment; and mitigate disadvantages of virtual teams (Jarvenpaa, 2004; Jarvenpaa & Leidner, 1999). Cognitive func-
 394 HOCH AND KOZLOWSKI

tioning can be represented in terms of team learning (Edmondson, the team level (i.e., team referents using a reference shift model of
1999; Edmondson, Bohmer, & Pisano, 2001), which is highly composition; Chan, 1998). Furthermore, all items were framed with
relevant due to the cognitively loaded nature of work in virtual respect to virtual team performance. The introduction of every page of
teams. For behavioral processes, member–member exchange qual- the questionnaire stated: “Please respond to all items with regard to
ity (Sherony & Green, 2002), which reflects the application of your work on your virtual team and not your regular line function,” or
traditional leader–member exchange (Gerstner & Day, 1997; “please respond to all of the following items with respect to your work
Graen, 1976; Graen & Uhl-Bien, 1995; Seers, 1989) to lateral on your virtual team.” Team members rated leadership and team
coworker exchange (i.e., among peers), is expected to be important composition. Team leaders rated the team’s performance. All of those
(Baltes, Dickson, Sherman, Bauer, & LaGanke, 2002; Hollings- items were measured using a 5-point Likert scale ranging from 1
head & McGrath, 1995). We expect that shared team leadership (strongly disagree) to 5 (strongly agree).
will provide a means to compensate for the gaps and disadvantages Hierarchical leadership was assessed using three indicators
in virtual teams such that it will be more strongly related to team rated by team members. Transformational leadership was mea-
performance with increasing levels of virtuality. Specifically, sured with the Multifactor Leadership Questionnaire (MLQ) 5X
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

(Avolio & Bass, 2004). Twenty items were used to measure

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Hypothesis 3: The positive relationship between shared team transformational leadership with its five subscales of attributed and
leadership (cognitive, affective, and behavioral leadership) behavioral idealized influence, intellectual stimulation, inspira-
and team performance increases as team virtuality increases. tional communication, and individual consideration. A sample
item for inspirational communication was “My team leader talks
Method optimistically about the future.” Cronbach’s ␣ was .92. Leader–
member exchange (LMX) was measured using a scale developed
by Graen, Novak, and Sommerkamp (1982; Wayne, Shore, &
Sample
Liden, 1997). A sample item was “My supervisor understands my
Study participants were comprised of 565 team members and problems and needs.” Cronbach’s ␣ was .89. Supervisor career
team leaders on 101 research and development (R & D) teams mentoring was measured by a validated scale (Blickle & Bou-
from global manufacturing industries. Human resource leaders in jataoui, 2005; Noe, 1988) based on the career support scale by
several companies were contacted. A number agreed on company Riley and Wrench (1985) that assessed three dimensions: career
participation in the study and facilitated data collection in ex- support, socio-emotional support, and role model. A sample item
change for technical report feedback and personal debriefing on is “My supervisor assigns tasks to me that foster the direct contact
the teams. The teams were similar in that all participants worked with important supervisors.” Cronbach’s ␣ was .89.
on R & D projects that involved knowledge-based, interdependent Structural supports were measured by (1) reward management and
group tasks. All teams worked under some degree of virtuality. (2) information and communication management, with one or two
That is, they worked across geographic distance, across different subscales each. Reward management was measured using a scale to
time zones, with employees from different cultural backgrounds, assess the quality of reward systems and a scale for fairness of reward
and primarily used electronic communication media for their work. systems. First, organizational reward management was measured
However, the degree of virtuality among the teams varied. While using five items from Podsakoff and MacKenzie (1994). A sample
some of them worked primarily face-to-face and to a limited item was “My performance appraisal system is very motivating.”
degree virtually, others ranged widely in the degree to which they Cronbach’s ␣ was .86. Second, a nine-item scale was used to
operated virtually. All of the teams primarily used electronic measure the fairness, transparency, and accountability of reward
communication media for their work, although some teams were systems based on Schminke et al. (2002). Specifically, items mea-
distributed across up to seven sites per team, whereas others pri- sured the three dimensions of performance evaluation, pay, and pro-
marily worked at one site. Team members on virtual teams worked on motion systems with regard to the extent they were perceived as (a)
average 359.20 miles away from each other, and 12% of the virtual fair, (b) accurate, and (c) transparent. Two sample items were “My
team members worked alone at one site. performance appraisal systems is fair,” “. . . transparent,” and so forth.
Teams consisted of an average of five members (SD ⫽ 2.94, Cronbach’s ␣ was .87.
range ⫽ 3–13). The average tenure of team members was 4.18 Information and communication management comprised five
years (SD ⫽ 4.96), and the average tenure of the leaders was items adapted from Clampitt and Downs (2004) assessing the
4.23 years (SD ⫽ 4.97). The average age of the team members was Information Quality, or Quality of Information Received. A sample
37 years (SD ⫽ 6.17, range ⫽ 19 – 61 years). Team members item was “Information that I receive is often unclear and not
averaged working on five projects at the same time (M ⫽ 4.86, precise” (R). Cronbach’s ␣ of this scale was .79. It also assessed
SD ⫽ 18.51). Since they were developing products together, their the extent to which Information Coordination Quality with a scale
work was interdependent. Therefore, task interdependence was adapted from Faraj and Sproull (2000). A sample item was: “There
measured as a control. Teams consisted of 77.1% male employees. is seldom confusion about how to accomplish our task.” Cron-
Average team leader age was 41 years (SD ⫽ 8.42, range ⫽ 25– 61 bach’s ␣ was .87.
years), and 89.1% of the team leaders were male. Shared leadership was measured in terms of cognitive, affec-
tive, and behavioral dimensions (Kozlowski & Bell, 2003; Koz-
lowski & Ilgen, 2006; Zaccaro et al., 2001). Specifically, cognitive
Measures
processes were measured with four items on team learning to
Scales. All constructs in the model center on the team as the assess the extent to which team members are active in obtaining
focal unit of theory. Accordingly, all measures were specified at feedback to improve their own performance. Here, a sample item
 LEADING VIRTUAL TEAMS 395

is “Our team actively searches our own performance for deficits.” 2003) on the individual level data to assess the fit of the measure-
Cronbach’s ␣ was .92. Next, affective correlates were measured ment model for the input factors and for the virtuality moderator
with five items on perceived team support (PTS), which had been composite. We accounted for the two-level hierarchical structure
developed on the basis of the perceived organizational support of the inputs, with the construct measures specified to load onto
construct by Eisenberger, Huntington, Hutchison, and Sowa their respective input variables. CFA also supported the two-level
(1986) and previously used by Wayne et al. (1997; Bishop, Scott, hierarchical structure of the virtuality moderator (i.e., geographical
& Burroughs, 2000). PTS measures the extent to which team distance combined with e-communication, and culture sub-
members support each other. A sample item is “My team really factors). Second, the main analyses were performed on group level
cares about my well-being.” Cronbach’s ␣ was .87. Finally, be- data. Given the reduction in sample size, we used partial least
havioral shared team leadership, in terms of member–member squares structural equation modeling (PLS), a regression-based
exchange (MMX) was measured with items that applied the leader– structural equation model (SEM) that is robust with regard to small
member exchange construct to the team consistent with Sherony samples (Chin, 2001; Ringle, Wende, & Will, 2006), which has
and Green (2002). Specifically, following a referent shift approach been adopted by many team researchers (Jung & Sosik, 2002;
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

(Chan, 1998), to measure MMX we referenced LMX-7 items (e.g., Sambamurthy & Chin, 1994; Sosik, Avolio, & Kahai, 1997). Tests
This document is copyrighted by the American Psychological Association or one of its allied publishers.

traditional measure of leader–member exchange; e.g., Gerstner & of significance in PLS were conducted using the bootstrap re-
Day, 1997; Graen & Uhl-Bien, 1995) to the team. A sample item sampling procedure (Efron & Tibshirani, 1993). Third, we tested
is “My team understands my problems and needs.” Cronbach’s ␣ moderation effects by computing the interaction terms between
was .87. team virtuality and the input variables using centered data follow-
Team performance was rated by team leaders on a scale based ing Aiken and West (1991).
on Hoegl and colleagues (Gemuenden & Hoegl, 2001; Hoegl & Justification for aggregation. The theoretical focus of the
Gemuenden, 2001). The team leader rated the team’s performance virtual team leadership model is specified at the team level. As-
regarding the aspects of work quantity, quality, keeping within the sessments of the input variables were obtained from individual
project schedule, and keeping within the budget using a scale team members using team referent items, which conforms to a
ranging from 0% to 100%. Cronbach’s ␣ was .79. referent shift composition model (Chan, 1998) for data aggrega-
Team virtuality. The degree of team virtuality was measured tion. Thus, we examined restricted within-group variance for all
in terms of three indicators: geographic dispersion, electronic variables prior to aggregation to the team level of analysis (Klein
communication media usage, and cultural differences (Fiol & et al., 2000; Kozlowski & Klein, 2000). We calculated ICC1,
O’Connor, 2005; Gibson & Cohen, 2003; Gibson & Gibbs, 2006; which is an index of inter-rater reliability, and ICC2, which is an
Kirkman & Malthieu, 2005; Kirkman et al., 2004; Townsend, index of the stability of the aggregated mean for each measure
DeMarie, & Hendrickson, 1998). Geographic dispersion was as- (Bliese, 2000). On average, across measures, the ICC1 was .45,
sessed with a measure by O’Leary & Cummings (2007), which and ICC2 was .69 (following r-to-z conversion). Specifically, the
included seven indicators, such as distance in miles, number of ICC1 and ICC2, respectively, were as follows: organizational
sites per team, percentage of team members alone at one site, and reward management was .44 and .75, reward systems was .44 and
others. The relative amount (frequency) of electronic versus face- .77, the quality of the information received was .46 and .67, the
to-face communication was measured with a scale based on Kac- way the knowledge was coordinated was .46 and .76, transforma-
mar, Witt, Zivnuska, and Gully (2003) that included indicators of tional leadership was .43 and .81, LMX was .43 and .83, mentoring
e-mail, chat, video and telephone conferencing, text and instant was .35 and .82, team learning was .43 and .80, perceived team
messaging, and face-to-face meetings that were rated with support (PTS) was .44 and .83, and member–member exchange
respect to frequency of use for communicating with colleagues (MMX) was .39 and .83. Overall, the ICC1 indices were substan-
and supervisors. Since we were interested in the relative fre- tial (Bliese, 2000), providing evidence to support aggregation, and
quency of electronic communication media use relative to total the ICC2 values indicated stability for the aggregated mean
communication, we calculated a ratio of relative communica- (Bliese, 2000).
tion frequency by dividing the sum of the electronic communi- Control variables. Since team age and gender composition
cation media use by the sum of all communication (media and correlated with several of the study variables, analyses were per-
face-to-face communication). To account for cultural differ- formed controlling for gender and age. We further controlled for
ences, we averaged the number of different nationalities per task interdependence (three items, based on Van Der Vegt, Emans,
team. The number of nationalities per teams on average was & Van De Vliert, 2000; Cronbach’s ␣ ⫽ .77) and the number of
3.60 (SD ⫽ 8.02) nationalities per team. projects an employee was working on (“How many projects are
The three scores of geographic distribution, electronic media you working on at the same time?”). We entered all five variables
usage, and cultural differences (nationalities per team) were sub- as controls into the PLS model.
ject to a z-transformation and were summed to form the team
virtuality composite. Cronbach’s ␣ was .77. The measure ranged Results
from –3.80 to 16.17, with M ⫽ – 0.32, and SD ⫽ 3.57. Higher
Means, standard deviations, inter-correlations, and reliability
scores indicate increased virtuality.
coefficients of study variables are presented in Table 1.

Analyses Pre-Analyses
Hypothesis testing was conducted in a three-step procedure. A CFA for the measurement model structure of the inputs was
First, we conducted confirmatory factor analyses (CFA; Arbuckle, performed. In order to determine if all the scales loaded on a single
 396 HOCH AND KOZLOWSKI

factor (representing a lack of distinction among the input factors),

.79
17
we first tested a one-factor model. This did not fit the data well
(␹2/df ⫽ 2.68, comparative fit index [CFI] ⫽ .89). Next, we

.92
.05
16

computed a three-factor model, where hierarchical leadership,

.50ⴱⴱ
15

structural supports, and shared team leadership were specified as

.87

.18
three separate constructs with all items loading on either hierar-

.63ⴱⴱ
.52ⴱⴱ
chical leadership, the structural supports, or the shared team lead-
14

.87

.09
ership construct. This three-factor model had a better fit (␹2/df ⫽
1.53, CFI ⫽ .96) and was a significant improvement compared to

.25ⴱⴱ
.26ⴱⴱ
.27ⴱⴱ
.89

⫺.09
the one-factor model, ⌬␹2(2) ⫽ 2,180.92. Analyses were con-
13

ducted using modification indices calculations (Buehner, 2004).

We then examined a hierarchically structured three-factor model,
.70ⴱⴱ
.26ⴱⴱ
.43ⴱⴱ
.29ⴱⴱ
12

.89

.13
where each item loaded first on their respective facet (i.e.,
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

transformational, LMX, mentoring, reward systems, extrinsic

This document is copyrighted by the American Psychological Association or one of its allied publishers.

.72ⴱⴱ
.68ⴱⴱ
.32ⴱⴱ
.39ⴱⴱ
.31ⴱⴱ
.92

⫺.12
motivation, information quality, information coordination, process
11

improvement, MMX, PTS). These 10 facets each formed the

second-order, hierarchical constructs for hierarchical leadership,
.28ⴱⴱ
.41ⴱⴱ

.29ⴱⴱ
.50ⴱⴱ
.44ⴱⴱ
.31ⴱⴱ
10

.87

.17

structural supports, or shared team leadership. The second-order

constructs of hierarchical leadership, structural supports, and
.49ⴱⴱ

.32ⴱⴱ
.21ⴱ

.24ⴱ

shared team leadership were allowed to intercorrelate. This

9

.79

.11
.03

.11
.19
Note. N ⫽ 101 teams. LMX ⫽ leader–member exchange; PTS ⫽ perceived team support; MMX ⫽ member–member exchange.

second-order, hierarchical model provided a good fit to the data

(␹2/df ⫽ 1.18, CFI ⫽ .99, root-mean-square error of approxima-
.26ⴱⴱ
.20ⴱ
.86
⫺.13
.06
.09

.04
.16
.13
.12
8

tion [RMSEA] ⫽ .02). Specifically, it demonstrated a better fit than

the previously tested three-factor, non-hierarchical model,
.51ⴱⴱ
.35ⴱⴱ
.28ⴱⴱ

.39ⴱⴱ
.41ⴱⴱ

⌬␹2(10) ⫽ 1,300.94, p ⬍ .001, and therefore provided the best fit

.24ⴱ

.22ⴱ
.23ⴱ
7

.87

.01

.09

for the theoretically expected factor structure. As a further indica-

tor for the quality of the measurement model, in PLS, the AVE-
.27ⴱⴱ
.20ⴱ

.23ⴱ
⫺.01
.13
.06

.16
⫺.04
.11
⫺.04
⫺.04
—
6

score (Fornell & Larcker, 1981) was calculated. The AVE-score

assesses the percentage of variance among the indicators caused by
⫺.33ⴱⴱ

the latent variable in relation the measurement error. The AVE-

.22ⴱ
⫺.05
⫺.16
⫺.11
⫺.19
.08
⫺.01

.08
⫺.06
.11
⫺.09
—
5

score should exceed .5. In the model we tested, the total AVE-score
was .85, which supports the quality of the measurement model.
.38ⴱⴱ
⫺.34ⴱⴱ

⫺.31ⴱⴱ
⫺.21ⴱ

The measurement model findings provide support for our approach

⫺.10
.05

⫺.16
⫺.18
.02
⫺.08
⫺.10
⫺.01
⫺.10
—
4

to use established constructs, validated measures, and composites

to represent the conceptual structure for the inputs.
.08
.05
⫺.02
⫺.10
.10
⫺.02
⫺.05
.10
.13
.12
.09
.10
⫺.02
⫺.18
—
3

We next used CFA to evaluate the structure of the virtuality

Means, Standard Deviations, and Pearson Correlation Coefficients

composite. Researchers have generally treated geographic distance

⫺.12
.03
⫺.14
.07
.01
.15
.01
⫺.08
⫺.03
⫺.07
⫺.10
.03
⫺.15
⫺.06
.11

and the use of electronic communication media as key components of

—
2

virtuality (e.g., Gibson & Cohen, 2003; Griffith et al., 2003; Kirkman
⫺.26ⴱⴱ
⫺.34ⴱⴱ

et al., 2004; Mesmer-Magnus et al., 2011; O’Leary & Cummings,

.22ⴱ
.04
⫺.10

.12
.02
.07
.11

⫺.13
⫺.02
⫺.17
.05
⫺.07
⫺.06
⫺.05
—
1

Gender: 1 ⫽ male; 2 ⫽ female. b Leader-rated variable.

2007; O’Leary & Mortensen, 2010). With the rise of globalization,

however, some theorists have made a conceptual case for incorporat-
6.16
0.35
2.97
8.02
3.70
1.30
0.74
0.96
0.58
0.71
0.58
0.67
0.67
0.52
0.59
0.76
9.41
SD

ing cultural differences into the conceptualization of team virtuality

(Hinds et al., 2011; Kirkman & Malthieu, 2005; Staples & Zhao,
37.22
1.33
5.36
3.60
0.14
3.59
2.88
2.71
2.42
3.60
3.21
3.93
2.79
3.14
3.80
3.20
89.45

2006; Tsui et al., 2007). Our composite conceptualization is consistent

M

with this emerging perspective. Thus, we used CFA to evaluate the

efficacy of this approach. We compared five models composed of the
Relative e-communication (Inv.)

three virtuality components:

1. a one-factor structure (i.e., all items loading on one

Geographic dispersion

factor);
Extrinsic motivation

Team Performanceb

p ⬍ .05. ⴱⴱ p ⬍ .01.
Information quality
Variable

Nationalities/team

Process feedback
Transformational
Reward systems

2. our composite—a one-factor hierarchical model with two

Coordination

sub-factors (i.e., electronic communication media com-

Mentoring
Team size
Gendera

bined with geographic dispersion; and nationality);

MMX
LMX
Table 1

PTS
Age

3. two separate factors (i.e., electronic media communication

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.

combined with geographic dispersion; and nationality);

ⴱ
a
 LEADING VIRTUAL TEAMS 397

4. a one-factor hierarchical model with three sub-factors (i.e., Table 3

electronic communication media, geographic dispersion, Specifications of the Inner Model Path Coefficients of the
and nationality) to represent separate contributions to a Distributed Leadership Model
higher-order factor; and
Team performance (supervisor rating)
5. three separate factors (i.e., electronic communication media, Predictor variables b SD t
geographic dispersion, and nationality).
Hierarchical leadership 0.85 0.13 0.67
As shown in Table 2, results indicated that Model 2, the hier- Structural supports 0.88ⴱⴱ 0.04 2.24
Shared team leadership 1.80ⴱⴱⴱ 0.04 2.23
archical one-factor model with two sub-factors (which reflected
our composite conceptualization) provided the best fit to the data. Gendera 0.96ⴱⴱⴱ 0.03 3.89
Model 1, the one-factor model with no sub-factors, did not fit the Age ⫺1.33ⴱ 0.04 3.39
data very well (␹2/df ⫽ 1.91, goodness-of-fit index [GFI] ⫽ .85, No. of projects ⫺3.01ⴱⴱⴱ 0.05 5.63
Task interdependence 0.96ⴱ 0.03 2.12
CFI ⫽ .93, RMSEA ⫽ .10). Fit for Model 2 (␹2/df ⫽ 1.39, GFI ⫽
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

.89, CFI ⫽ .97, RMSEA ⫽ .06) was good. Fit for Model 2 was
This document is copyrighted by the American Psychological Association or one of its allied publishers.

R2 .18
significantly better than for Model 1, the one-factor model,
Note. N ⫽ 101 teams; Bootstrap 500: 2000.
⌬␹2(3) ⫽ 48.27, p ⬍ .001, or for Model 3, the two-factor non- a
Gender: 1 ⫽ male; 2 ⫽ female.
hierarchical model, ⌬␹2(1) ⫽ 23.85, p ⬍ .001. Model 2 also fit the ⴱ
p ⬍ .05. ⴱⴱ p ⬍ .01. ⴱⴱⴱ p ⬍ .001.
data better than Model 5, the three-factor model, ⌬␹2(14) ⫽ 29.58,
p ⬍ .01, which showed poor fit overall (␹2/df ⫽ 1.85, GFI ⫽ .84,
CFI ⫽ .93, RMSEA ⫽ .09). Thus, there was good empirical
significant main relationship with team performance (b ⫽ 0.55, ns)
support for our conceptualization and composite approach to cap-
when team virtuality was added, but structural supports interacted
turing team virtuality.
with virtuality in predicting team performance (b ⫽ 2.77, p ⬍
.001). Those results are displayed in Table 4.
Main Analyses: Inner Model Analyses As shown in Figures 2 and Figures 3, we graphed these rela-
Next, we examined direct relationships via the inner (structural tionships following Aiken and West (1991). Figure 1 shows that
model analyses) from PLS with the three groups of predictor under high levels of virtuality hierarchical leadership was not
variables predicting team effectiveness on the team level data. The related to team performance. Under low levels of virtuality, hier-
inner model analyses showed that both structural supports (b ⫽ archical leadership was significantly related to team performance.
0.88, p ⬍ .01) and shared team leadership (b ⫽ 1.80, p ⬍ .001) Figure 2 shows that structural supports were positively related to
predicted team performance, whereas hierarchical leadership did team performance under high levels of virtuality, but not under low
not (b ⫽ 0.85, ns). Those results are displayed in Table 3. virtuality. Hypotheses 1 and 2 were supported. Finally, for Hy-
pothesis 3, team virtuality did not interact with shared leadership
in predicting team performance (ns), whereas shared team leader-
Moderation by Team Virtuality
ship was still positively related to team performance (b ⫽ 1.94,
Moderation analyses were conducted to determine whether the p ⬍ .001). Hypothesis 3 was not supported; results are displayed
degree of team virtuality had a differential influence on the rela- in Table 4.
tionship between the inputs and team performance as predicted by
the model. Centered data were used to compute the interaction Discussion
terms between team virtuality and the three groups of predictor
variables. When team virtuality was entered as a predictor, there Summary
was a marginally negative relationship with team performance
(b ⫽ – 0.40, p ⬍ .10), suggesting that with increasing levels of This study examined the relationships between hierarchical
team virtuality teams performed less well. leadership, structural supports, and shared team leadership with
Next, hierarchical leadership interacted with team virtuality in team performance, and the moderating effects of virtuality on these
predicting team performance in a negative way (b ⫽ –1.74, p ⬍ relationships. Our research approach, which assessed each of the
.01). With regard to structural supports, there was no longer a inputs as construct composites, provided a measurement model

Table 2
Model Comparisons for Measurement of Virtuality

Model ␹2/df GFI CFI RMSEA

1. One-factor model (without sub-factors) 1.91 .85 .93 .10

2. One-factor model with two sub-factors (hierarchical two-factor model) 1.39 .89 .97 .06
3. Two-factor model (non-hierarchical) 1.66 .87 .95 .08
4. One-factor model with three sub-factors (hierarchical three-factor model) 1.85 .84 .93 .09
5. Three-factor model (non-hierarchical) 1.85 .84 .93 .09
Note. N ⫽ 101. GFI ⫽ goodness-of-fit index; CFI ⫽ comparative fit index; RMSEA ⫽ root-mean-square error of approximation.
 398 HOCH AND KOZLOWSKI

Table 4

Team Performance (In %)

Moderation of Leadership Variables’ Effects by Team Virtuality
on Team Outcomes: Specification of Inner Model Path
Coefficients of the Distributed Leadership Model
low Virtuality
Team performance
(supervisor rating) high Virtuality
Predictor b SD t

Hierarchical leadership ⫺0.06 0.09 0.07

Hierarchical Leadership ⫻ Team Virtuality ⫺1.74ⴱⴱ 0.07 2.32 low Structural high Structural
Structural supports 0.55 0.05 0.75 Supports Supports
Structural Supports ⫻ Team Virtuality 2.77ⴱⴱⴱ 0.06 4.34
Shared team leadership 1.94ⴱⴱⴱ 0.06 3.37
Shared Team Leadership ⫻ Team Virtuality 0.70 0.09 0.78
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Team virtuality ⫺0.40† 0.07 0.89 Figure 3. Interaction between structural supports and team virtuality
This document is copyrighted by the American Psychological Association or one of its allied publishers.

predicting team performance.

Gendera 1.16ⴱⴱⴱ 0.03 3.21
Age ⫺1.59ⴱⴱ 0.05 3.19
No. of projects ⫺3.01ⴱⴱⴱ 0.06 5.45 media usage (Griffith et al., 2003; Kirkman et al., 2004), and
Task interdependence 0.39 0.04 0.89 cultural diversity (Hinds et al., 2011). With regard to cultural
diversity, our work reflects recent theorizing in the literature that
R2 .27
cultural differences add to virtuality (e.g., Chen et al., 2010; Hinds
Note. N ⫽ 101 teams; Bootstrap 500: 2,000 teams. et al., 2011; Tsui et al., 2007).
a
Gender: 1 ⫽ male; 2 ⫽ female. Leadership in virtual teams. The main goal of the present
†
p ⬍ .10. ⴱⴱ p ⬍ .01. ⴱⴱⴱ p ⬍ .001.
research was to investigate hierarchical leadership in teams, the
inhibitory impact of virtuality on hierarchical leadership, and the
that exhibited a good fit to the data in our sample. With regard ability of structural supports and shared team leadership supple-
direct relations, structural supports and shared leadership, but not ment it. Our findings showed that the influence of hierarchical
hierarchical leadership, were positively associated with team per- leadership on team performance is weakened when teams are more
formance. When testing for moderating effects, structural supports virtual in nature. Thus, when teams are virtual, it is desirable to
were more, and hierarchical leadership was less, strongly associ- supplement the leader behaviors that are mitigated by distance,
ated with team performance, the higher the level of team virtuality. electronic media, and cultural differences. Following Bell and
The association between shared team leadership and team perfor- Kozlowski (2002), we examined the role of structural support
mance was not affected by the degree of virtuality. mechanisms and shared team leadership as alternative inputs to
team performance that could mitigate the loss of influence. Our
findings show that hierarchical leadership had weaker relations,
Theoretical Implications
whereas structural supports were more strongly related, with
The conceptualization and measurement of team virtuality. team performance under increasing levels of team virtuality.
First, central to virtual team leadership is the need to examine the Contrary to expectations, shared team leadership exhibited sta-
appropriate form of measurement to capture the multifaceted con- ble positive relations with team performance regardless of the
cept of team virtuality. This is a primary contribution of our degree of virtuality.
research. The measure we developed incorporated the most recent
theorizing regarding the underpinnings of the virtuality construct; Limitations and Research Extensions
it included geographic dispersion (O’Leary & Mortensen, 2010),
the relative amount of face-to-face and electronic communication Theorists have speculated that the processes of hierarchical
leadership are disadvantaged under conditions of virtuality, and
that supplements by structural supports and shared team leadership
can mitigate the loss of leadership influence (e.g., Bell & Kozlow-
Team Performance (In %)

ski, 2002). Our approach to examining this speculation was delib-

erately focused; we examined these factors as inputs to indepen-
dently rated team performance and treated them as construct
low Virtuality composites. Although this allowed a parsimonious examination of
high Virtuality the basic relationships of interest, it does not address the underly-
ing process mechanisms by which hierarchical leadership is inhib-
ited and structural supports provide supplements. Moreover, given
low Hierarchical high Hierarchical that shared team leadership exhibited a consistent relationship with
Leadership Leadership team performance regardless of the degree of virtuality, unpacking
the mechanisms by which it manifests this broad influence is
clearly in order.
Figure 2. Interaction between hierarchical leadership and team virtuality There are some considerations that should be addressed in future
predicting team performance. research to extend these findings. First, with moderation of the
 LEADING VIRTUAL TEAMS 399

input factors to performance relationship by virtuality established, tion, and/or training could be potentially useful. They might also
the next increment should turn attention to the mediating mecha- need more time and resources for leading their virtual team com-
nisms that link the input factors with team performance. This will pared to leading their respective face-to-face team. Second, struc-
necessitate longitudinal research designs to appropriately capture tural supports are more strongly related to team performance in
the processes and minimize concerns about causal ambiguity. more virtual teams. Thus, structural supports have the potential to
Second, although common source method variance is not an issue be an effective management tool for augmenting hierarchical lead-
with respect to the relationship between the input factors and ership and can be recommended to aid leaders managing virtual
performance (which was rated independently), it will be desirable teams. Structural supports comprise fair and reliable reward sys-
to distinguish inputs from mediating processes in future research. tems, and transparent communication and information manage-
This may be accomplished by cross-splitting teams to examine ment. Based on our findings, structural supports should be imple-
relations between inputs and processes (e.g., Hofmann & Stetzner, mented to augment hierarchical leaders in virtual teams. Third,
1996), although such designs necessitate large teams. With basic although expected, shared leadership contributed to team perfor-
input-output-moderation relations established, mediating pro- mance regardless of the degree of virtuality. Therefore, shared
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

cesses are obvious next steps for extension. leadership can be recommended for the management of all teams
This document is copyrighted by the American Psychological Association or one of its allied publishers.

In addition, there are also issues surrounding the composite along the virtuality continuum.
approach used to capture the degree of team virtuality—which has Based on our findings, structural supports can be recommended
both advantages and disadvantages—that merit discussion. With for managing virtual teams and shared leadership can be recom-
respect to advantages, the composite—which combines the facets mended for managing teams in general. With regard to structural
of geographical separation, use of electronic media, and cultural supports, future research should determine the extent to which
diversity—is consistent with the conceptual evolution of the con- leaders, or others in the organization, could influence perceptions
cept of virtuality that has occurred over the last decade. Estab- of structural supports among virtual team members. For example,
lished conceptualizations of virtuality focus on geographical sep- high structural supports might be less salient when there are
aration and the use of electronic media (Gibson & Cohen, 2003; restrictions in technology or resources, when reward systems do
Griffith et al., 2003; Kirkman et al., 2004; Mesmer-Magnus et al., not reward team performance (or any performance!), or when there
2011; O’Leary & Cummings, 2007; O’Leary & Mortensen, 2010). are low levels of organizational support. Another direction for
However, organizations have become increasingly multi-national, future research is to more systematically investigate the boundary
work teams are more frequently dispersed around the world, and conditions of structural supports, as well as moderating variables
technological interconnectivity continues to advance. By treating that might influence the effectiveness of structural supports.
the facets as a composite, we captured a richer conceptualization of Shared team leadership enhanced team performance regardless
virtuality in a parsimonious fashion that was also empirically of virtuality. This was unexpected, as the literature has viewed this
supported in our data. supplement as more important under greater degrees of virtuality.
This conceptualization and assessment, however, also intro- This study extends prior literature with regard to the conception of
duces ambiguity with respect to the definition of virtuality and to shared leadership as a means to supplement team functions, as it
the precise contribution of the distinct components. Each compo- captures the extent to which team members can collectively en-
nent is a unique characteristic, and it could be argued that their gage in cognitive, affective/motivational, and behavioral team
combination, while richer, is also less precise. Clearly, there is leadership behaviors (Kozlowski & Bell, 2003).
value in identifying the unique influences of the specific compo- Because the influence of shared leadership on team performance
nents of virtuality that we combined. Examining the effects of each was not affected by degrees of virtuality, shared team leadership
component as a distinct moderator, in combination with the other appears to have the potential to be a potent leadership approach.
components, however, will necessitate sampling that can achieve However, there is a lack of research focused on the antecedents of
wide variance on each component and substantial sample sizes to shared team leadership (e.g., Carson et al., 2007; Hoch, in press).
allow robust evaluations. These sampling issues would be com- This has implications for theoretical extension and practical appli-
pounded as additional components of virtuality are proposed. This cation: the question of how to facilitate the emergence of shared
is desirable research extension, although we acknowledge that leadership has not been addressed by prior research or this study
such data will be challenging to acquire. (e.g., Pearce & Conger, 2003). Future research needs to identify
Finally, generalization of research findings is always limited by antecedents, mediators, and moderators of shared team leadership,
the nature of the sample. Our teams were engaged in research and such as the impact of self-leadership and self-management (Manz,
development activities and drawn from a diverse set of firms in the 1986; Neck & Houghton, 2006), group potency, group self-
global automotive and automotive supplies industry. Clearly there efficacy, and team cohesion (Bandura, 1997; Chen, Gully, & Eden,
is a need to replicate the findings in teams that work in other 2001; Guzzo & Shea, 1992) or team and task conditions on shared
contexts, industries, and cultural settings. team leadership effects.

Practical Implications and Future Research Conclusion

There are three main practical implications. First, our data The nature of global competition and continuing advances in
suggest that the influence of hierarchical leadership is mitigated in communication technologies means that virtual teams are an inte-
virtual teams, as it is less strongly related to team performance gral aspect of work structure worldwide. They challenge what we
when teams are more virtual in nature. This finding suggests that know and need to know about leading and managing teams. We
providing virtual team leaders with appropriate support, orienta- hope that these findings help to advance additional research on the
 400 HOCH AND KOZLOWSKI

role of leadership, and leadership supplements, for enhancing team Chan, D. (1998). Functional relations among constructs in the same content
performance across the range of team virtuality. domain at different levels of analysis: A typology of composition mod-
els. Journal of Applied Psychology, 83, 234 –246. doi:10.1037/0021-
9010.83.2.234
References Chao, G. T., Walz, P., & Gardner, P. D. (1992). Formal and informal
mentorships: A comparison on mentoring functions and contrast with
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and
nonmentored counterparts. Personnel Psychology, 45, 619 – 636.
interpreting interactions. Newbury Park, CA: Sage.
Chen, G., Gully, C. S., & Eden, D. (2001). Validation of a new general
Allen, T. D., Eby, L. T., Poteet, M. L., Lentz, E., & Lima, L. (2004). Career
self-efficacy scale. Organizational Research Methods, 4, 62– 83. doi:
benefits associated with mentoring for proteges: A meta-analysis. Jour-
nal of Applied Psychology, 89, 127–136. doi:10.1037/0021-9010.89.1 10.1177/109442810141004
.127 Chen, G., Kirkman, B. L., Kim, K. H., Farh, C. I. C., & Tangirala, S.
Arbuckle, J. L. (2003). AMOS 5.0. Chicago, IL: Smallwaters. (2010). When does cross-cultural motivation enhance expatriate effec-
Avolio, B. J., & Bass, B. M. (2004). MLQ—Multifactor Leadership Ques- tiveness? A multilevel investigation of the moderating roles of subsid-
tionnaire. Menlo Park, CA: Mind Garden. iary support and cultural distance. Academy of Management Journal, 53,
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Avolio, B. J., Jung, D. I., Murry, W., & Sivasbramaniam, N. (1996). 1110 –1130. doi:10.5465/AMJ.2010.54533217
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Building highly developed teams: Focusing on shared leadership pro- Cheshin, A., Rafaeli, A., & Bos, N. (2011). Anger and happiness in virtual
cesses, efficacy, trust, and performance. In M. M. Beyerlein, S. Beyer- teams: Emotional influences of text and behavior on others’ affect in the
lein, & D. A. Johnson (Eds.), Advances in interdisciplinary studies of absence of non-verbal cues. Organizational Behavior and Human De-
work teams: Team leadership (pp. 173–209). Greenwich, CT: JAI Press. cision Processes, 116, 2–16. doi:10.1016/j.obhdp.2011.06.002
Avolio, B. J., & Kahai, S. S. (2003). Adding the “E” to e-leadership: How Chin, W. W. (2001). PLS-graph user’s guide: Version 3.0. Houston, TX:
it may impact your leadership. Organizational Dynamics, 31, 325–338. University of Houston, C. T. Bauer College of Business.
doi:10.1016/S0090-2616(02)00133-X Clampitt, P. G., & Downs, C. W. (2004). Down–Hazen Communication
Avolio, B. J., Kahai, S. S., & Dodge, G. E. (2000). E-leadership: Impli- Satisfaction Questionnaire. In C. W. Downs & A. D. Adrian (Eds.),
cations for theory, research, and practice. The Leadership Quarterly, 11, Assessing organizational communication (pp. 139 –157). New York,
615– 668. doi:10.1016/S1048-9843(00)00062-X NY: Guilford Press.
Baltes, B. B., Dickson, M. W., Sherman, M. P., Bauer, C. C., & LaGanke, Colquitt, J. A. (2004). Does the justice of the one interact with the justice
J. S. (2002). Computer mediated communication and group decision of the many? Reactions to procedural justice in teams. Journal of
making: A meta-analysis. Organizational Behavior and Human Deci-
Applied Psychology, 89, 633– 646. doi:10.1037/0021-9010.89.4.633
sion Processes, 87, 156 –179. doi:10.1006/obhd.2001.2961
Day, D. V., Gronn, P., & Salas, E. (2004). Leadership capacity in teams.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York, NY:
The Leadership Quarterly, 15, 857– 880. doi:10.1016/j.leaqua.2004.09
Freeman.
.001
Bass, B. M. (1985). Leadership and performance beyond expectations.
New York, NY: Free Press. Duarte, D. L., & Snyder, N. T. (2001). Mastering virtual teams (2nd ed.).
Bass, B. M. (1998). Transformational leadership: Industrial, military, and San Francisco, CA: Jossey-Bass.
educational impact. Hillsdale, NJ: Erlbaum. Dulebohn, J. H., & Martocchio, J. J. (1998). Employee perceptions of the
Bell, B. S., & Kozlowski, S. W. J. (2002). A typology of virtual teams: fairness of work group incentive plans. Journal of Management, 24,
Implications for effective leadership. Group & Organization Manage- 469 – 488.
ment, 27, 14 – 49. doi:10.1177/1059601102027001003 Edmondson, A. (1999). Psychological safety and learning behavior in work
Bishop, J. W., Scott, K. D., & Burroughs, S. M. (2000). Support, commit- teams. Administrative Science Quarterly, 44, 350 –383. doi:10.2307/
ment, and employee outcomes in a team environment. Journal of Man- 2666999
agement, 26, 1113–1132. doi:10.1177/014920630002600603 Edmondson, A., Bohmer, R. M., & Pisano, G. P. (2001). Disrupted
Blickle, G., & Boujataoui, M. (2005). Mentoren, Karriere und Geschlecht: routines: Team learning and new technology implementation in hospi-
Eine Feldstudie mit Führungskräften aus dem Personalbereich [Mentors, tals. Administrative Science Quarterly, 46, 685–716. doi:10.2307/
career, and gender: A field study with supervisors from personnel 3094828
departments]. Zeitschrift für Arbeits- und Organisationspsychologie, 49, Efron, B., & Tibshirani, R. J. (1993). An introduction to the bootstrap. New
1–11. doi:10.1026/0932-4089.49.1.1 York, NY: Chapman & Hall/CRC.
Bliese, P. D. (2000). Within-group agreement, non-interdependence, and Eisenberger, R., Huntington, R., Hutchison, S., & Sowa, D. (1986). Per-
reliability—Implications for data aggregation and analysis. In K. J. Klein ceived organizational support. Journal of Applied Psychology, 71, 500 –
& S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methods 507. doi:10.1037/0021-9010.71.3.500
in organizations: Foundations, extensions and new directions (pp. 349 – Ensley, M. D., Hmieleski, K. M., & Pearce, C. L. (2006). The importance
382). San Francisco, CA: Jossey-Bass.
of vertical and shared leadership within new venture top management
Bowers, D. G., & Seashore, S. E. (1966). Predicting organizational effec-
teams: Implications for the performance of startups. The Leadership
tiveness with a four-factor theory of leadership. Administrative Science
Quarterly, 17, 217–231. doi:10.1016/j.leaqua.2006.02.002
Quarterly, 11, 238 –263. doi:10.2307/2391247
Faraj, S., & Sproull, L. (2000). Coordinating expertise in software devel-
Brown, M. E., & Gioia, D. A. (2002). Making things click: Distributive
leadership in an online division of an offline organization. The Leader- opment teams. Management Science, 46, 1554 –1568. doi:10.1287/mnsc
ship Quarterly, 13, 397– 419. doi:10.1016/S1048-9843(02)00123-6 .46.12.1554.12072
Buehner, M. (2004). Einführung in die Test- und Fragebogenkonstruktion- Fiol, C. M., & O’Connor, E. J. (2005). Identification in face-to-face,
[Introduction into Test and Questionnaire-Construction]. Munich, Ger- hybrid, and pure virtual teams: Untangling the contradictions. Organi-
many: Pearson. zation Science, 16, 19 –32. doi:10.1287/orsc.1040.0101
Carson, J. B., Tesluk, P. E., & Marrone, J. A. (2007). Shared leadership in Fleishman, E. A., Mumford, M. D., Zaccaro, S. J., Levin, K. Y., Korotkin,
teams: An investigation of antecedent conditions and performance. A. L., & Hein, M. B. (1991). Taxonomic efforts in the description of
Academy of Management Journal, 50, 1217–1234. doi:10.2307/ leader behavior: A synthesis and functional interpretation. The Leader-
20159921 ship Quarterly, 2, 245–287. doi:10.1016/1048-9843(91)90016-U
 LEADING VIRTUAL TEAMS 401

Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models Hofmann, D. A., & Stetzer, A. (1996). A cross-level investigation of
with unobservable variables and measurement error. Journal of Market- factors influencing unsafe behaviors and accidents. Personnel Psychol-
ing Research, 18, 39 –50. doi:10.2307/3151312 ogy, 49, 307–339. doi:10.1111/j.1744-6570.1996.tb01802.x
Fuller, J. B., Patterson, C. E. P., Hester, K., & Stringer, D. Y. (1996). A Hollingshead, A. B., & McGrath, J. E. (1995). Computer-assisted groups:
quantitative review of research on charismatic leadership. Psychological A critical review of the empirical research. In R. A. Guzzo & E. Salas
Reports, 78, 271–287. doi:10.2466/pr0.1996.78.1.271 (Eds.), Team effectiveness and decision making in organizations (pp.
Gemuenden, H. G., & Hoegl, M. (2001). Management von Teams: Theo- 46 –78). San Francisco, CA: Jossey-Bass.
retische Konzepte und empirische Befunde [Management of teams, the- Howell, J. M., & Hall-Merenda, K. (1999). The ties that bind: The impact
oretical concepts, and empirical results]. Wiesbaden, Germany: Gabler. of leader–member exchange, transformational and transactional leader-
Gerstner, C. R., & Day, D. V. (1997). Meta-analytic review of leader– ship, and distance on predicting follower performance. Journal of Ap-
member exchange theory: Correlates and construct issues. Journal of plied Psychology, 84, 680 – 694. doi:10.1037/0021-9010.84.5.680
Applied Psychology, 82, 827– 844. doi:10.1037/0021-9010.82.6.827 Howell, J. M., Neufeld, D. J., & Avolio, B. J. (2005). Examining the
relationship of leadership and physical distance with business unit per-
Gibson, C. B., & Cohen, S. G. (2003). Virtual teams that work: Creating
formance. The Leadership Quarterly, 16, 273–285. doi:10.1016/j.leaqua
conditions for virtual team effectiveness. San Francisco, CA: Jossey-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

.2005.01.004
Bass.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Howell, J. P., & Dorfman, P. W. (1981). Substitutes for leadership: Test of

Gibson, C. B., & Gibbs, J. L. (2006). Unpacking the concept of virtuality:
a construct. Academy of Management Journal, 24, 714 –728. doi:
The effects of geographic dispersion, electronic dependence, dynamic
10.2307/256171
structure, and national diversity on team innovation. Administrative
Howell, J. P., & Dorfman, P. W. (1986). Leadership and substitutes for
Science Quarterly, 51, 451– 495.
leadership among professional and non-professional workers. Jour-
Graen, G. B. (1976). Role-making processes within complex organizations. nal of Applied Behavioral Science, 22, 29 – 46. doi:10.1177/
In M. D. Donette (Ed.), Handbook of industrial and organizational 002188638602200106
psychology (pp. 1201–1245). Chicago, IL: Rand-McNally. i4cp. (2006, May 5). Beyond virtual teams. Retrieved from http://www
Graen, G., Novak, M., & Sommerkamp, P. (1982). The effects of leader– .i4cp.com/trendwatchers/2006/05/05/beyond-virtual-teams
member exchange and job design on productivity and satisfaction: i4cp. (2008, September 4). Virtual teams now a reality: Two out of three
Testing a dual attachment model. Organizational Behavior and Human companies say they will rely more on virtual teams in the future.
Performance, 30, 109 –131. doi:10.1016/0030-5073(82)90236-7 Retrieved from http://www.i4cp.com/news/2008/09/04/virtual-
Graen, G., & Uhl-Bien, M. (1995). Relationship-based approach to lead- teams-now-a-reality
ership: Development of leader–member exchange (LMX) theory of Irwin, J. R., & McClelland, G. H. (2003). Negative consequences of
leadership over 25 years: Applying a multi-level multi-domain perspec- dichotomizing continuous predictor variables. Journal of Market Re-
tive. The Leadership Quarterly, 6, 219 –247. doi:10.1016/1048- search, 40, 366 –371. doi:10.1509/jmkr.40.3.366.19237
9843(95)90036-5 Jarvenpaa, S. L. (2004). Toward conceptualized theories of trust: The role
Griffith, T. L., Sawyer, J. E., & Neale, M. A. (2003). Virtualness and of trust in global virtual teams. Information Systems Research, 15,
knowledge in teams: Managing the love triangle of organizations, indi- 250 –267. doi:10.1287/isre.1040.0028
viduals, and information technology. MIS Quarterly, 27, 265–287. Jarvenpaa, S. L., & Leidner, D. E. (1999). Communication and trust in
Guzzo, R. A., & Shea, G. P. (1992). Group performance and intergroup global virtual teams. Organization Science, 10, 791– 815. doi:10.1287/
relations in organizations. Palo Alto, CA: Consulting Psychologists orsc.10.6.791
Press. Judge, T. A., & Piccolo, R. F. (2004). Transformational and transactional
Hambley, L. A., O’Neill, T. A., & Kline, T. J. B. (2007). Virtual team leadership: A meta-analytic test of their relative validity. Journal of
leadership: The effects of leadership style and communication medium Applied Psychology, 89, 755–768. doi:10.1037/0021-9010.89.5.755
on team interaction styles and outcomes. Organizational Behavior and Jung, D. I., & Sosik, J. J. (2002). Transformational leadership in work
Human Decision Processes, 103, 1–20. doi:10.1016/j.obhdp.2006.09 groups: The role of empowerment, cohesiveness, and collective-efficacy
.004 on perceived group performance. Small Group Research, 33, 313–336.
Hamilton, B. A., & Scandura, T. A. (2003). E-mentoring: Implications for doi:10.1177/10496402033003002
Kacmar, K. M., Witt, L. A., Zivnuska, S., & Gully, S. M. (2003). The
organizational learning and development in a wired world. Organiza-
interactive effect of leader–member exchange and communication fre-
tional Dynamics, 31, 388 – 402. doi:10.1016/S0090-2616(02)00128-6
quency on performance ratings. Journal of Applied Psychology, 88,
Hill, S. N., Bartol, K. M., Tesluk, P. E., & Langa, G. A. (2009). Organi-
764 –772. doi:10.1037/0021-9010.88.4.764
zational context and face-to-face interaction: Influences on the develop-
Kahai, S. S., Sosik, J. J., & Avolio, B. J. (2003). Effects of leadership style,
ment of trust and collaborative behaviors in computer-mediated groups.
anonymity, and rewards on creativity-relevant processes and outcomes
Organizational Behavior and Human Decision Processes, 108, 187–
in an electronic meeting system context. The Leadership Quarterly, 14,
201. doi:10.1016/j.obhdp.2008.10.002
499 –524. doi:10.1016/S1048-9843(03)00049-3
Hinds, P., & Kiesler, S. (2002). Distributed work. Cambridge, MA: MIT Kanawattanachai, P., & Yoo, Y. (2007). The impact of knowledge coor-
Press. dination on virtual team performance over time. MIS Quarterly, 31,
Hinds, P., Liu, L., & Lyon, J. (2011). Putting the global in global work: An 783– 808.
intercultural lens on the practice of cross-national collaboration. The Kasper-Fuehrer, E. C., & Ashkanasy, N. M. (2001). Communicating trust-
Academy of Management Annals, 5, 135–188. doi:10.1080/19416520 worthiness and building trust in inter-organizational virtual organiza-
.2011.586108 tions. Journal of Management, 27, 235–254.
Hoch, J. E. (in press). Shared leadership and innovation: The role of Kerr, S. (1977). Substitutes for leadership: Some implications for organi-
vertical leadership and employee integrity. Journal of Business and zational design. Organization and Administrative Sciences, 8, 135–146.
Psychology. Kerr, S., & Jermier, J. M. (1978). Substitutes for leadership: Their meaning
Hoegl, M., & Gemuenden, H. G. (2001). Teamwork quality and the success and measurement. Organizational Behavior and Human Performance,
of innovative projects: A theoretical concept and empirical evidence. 22, 375– 403. doi:10.1016/0030-5073(78)90023-5
Organization Science, 12, 435– 449. doi:10.1287/orsc.12.4.435.10635 Kiesler, S., & Cummings, J. N. (2002). What do we know about proximity
 402 HOCH AND KOZLOWSKI

and distance in work groups? A legacy of research.In P. Hinds & S. Martins, L. L., & Shalley, C. E. (2011). Creativity in virtual work: Effects
Kiesler (Eds.), Distributed work (pp. 57– 80). Cambridge, MA: MIT of demographic differences. Small Group Research, 42, 536 –561. doi:
Press. 10.1177/1046496410397382
Kirkman, B. L., Gibson, C. B., & Kim, K. (2012). Across borders and Mayo, M., Meindl, J. R., & Pastor, J. C. (2003). Shared leadership in work
technologies: Advancements in virtual teams research. In S. W. J. teams. In C. L. Pearce & J. A. Conger (Eds.), Shared leadership:
Kozlowski (Ed.), Oxford handbook of organizational psychology (Vol. Reframing the hows and whys of leadership (pp. 193–214). doi:10.4135/
2, pp. 789 – 858). New York, NY: Oxford University Press. 9781452229539.n9
Kirkman, B. L., & Malthieu, J. E. (2005). The dimensions and antecedents McGrath, J. E. (1962). Leadership behavior: Some requirements for lead-
of team virtuality. Journal of Management, 31, 700 –718. doi:10.1177/ ership training. Washington, DC: U.S. Civil Service Commission, Of-
0149206305279113 fice of Career Development.
Kirkman, B. L., Rosen, B., Tesluk, P. E., & Gibson, C. B. (2004). The Mehra, A., Smith, B. R., Dixon, A. L., & Robertson, B. (2006). Distributed
impact of team empowerment on virtual team performance: The mod- leadership in teams: The network of leadership perceptions and team
performance. The Leadership Quarterly, 17, 232–245. doi:10.1016/j
erating role of face-to-face-interaction. Academy of Management Jour-
.leaqua.2006.02.003
nal, 47, 175–192. doi:10.2307/20159571
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Mesmer-Magnus, J. R., DeChurch, L. A., Jimenez-Rodriguez, M., Wild-

Klein, K. J., Bliese, P. D., Kozolowski, S. W. J., Dansereau, F., Gavin,
This document is copyrighted by the American Psychological Association or one of its allied publishers.

man, J., & Shuffler, M. (2011). A meta-analytic investigation of virtu-

M. B., Griffin, M. A., . . . Bligh, M. C. (2000). Multilevel analytical
ality and information sharing in teams. Organizational Behavior and
techniques: Commonalities, differences, and continuing questions. In
Human Decision Processes, 115, 214 –225. doi:10.1016/j.obhdp.2011
K. J. Klein & S. W. J. Kozlowski (Eds.), Multilevel theory, research, and
.03.002
methods in organizations: Foundations, extensions, and new directions
Morgeson, F. P., DeRue, D. S., & Karam, E. P. (2010). Leadership in
(pp. 512–553). San Francisco, CA: Jossey-Bass. teams: A functional approach to understanding leadership structures and
Kozlowski, S. W. J., & Bell, B. S. (2003). Work groups and teams in processes. Journal of Management, 36, 5–39. doi:10.1177/
organizations. In W. C. Borman, D. R. Ilgen, & R. J. Klimoski (Eds.), 0149206309347376
Comprehensive handbook of psychology: Industrial and organizational Neck, C. P., & Houghton, J. D. (2006). Two decades of self-leadership
psychology (pp. 333–375). New York, NY: Wiley. theory and research. Journal of Managerial Psychology, 21, 270 –295.
Kozlowski, S. W. J., & Ilgen, D. R. (2006). Enhancing the effectiveness of doi:10.1108/02683940610663097
work groups and teams. Psychological Science in the Public Interest, 7, Noe, R. A. (1988). An investigation of the determinants of successful
77–124. assigned mentoring relationships. Personnel Psychology, 41, 457– 479.
Kozlowski, S. W. J., & Klein, K. J. (2000). A multilevel approach to theory doi:10.1111/j.1744-6570.1988.tb00638.x
and research in organizations: Contextual, temporal, and emergent pro- O’Leary, M. B., & Cummings, J. N. (2007). The spatial, temporal, and
cesses. In K. J. Klein & S. J. Kozlowski (Eds.), Multilevel theory, configurational characteristics of geographic dispersion in teams. MIS
research, and methods in organizations: Foundations, extensions, and Quarterly, 31, 433– 452.
new directions (pp. 3–90). San Francisco, CA: Jossey-Bass. O’Leary, M. B., & Mortensen, M. (2010). Go con(figure): Subgroups,
Kram, K. E. (1985). Mentoring at work. Glenview, IL: Scott Foresman and imbalance, and isolates in geographically dispersed teams. Organization
Company. Science, 21, 115–131. doi:10.1287/orsc.1090.0434
Künzle, B., Zala-Mezö, E., Wacker, J., Kolbe, M., Spahn, D. R., & Grote, Ostroff, C., & Kozlowski, S. W. J. (1993). The role of mentoring in the
G. (2010). Leadership in anaesthesia teams: The most effective leader- information gathering processes of newcomers during early organiza-
ship is shared. Quality and Safety Health Care, 19, 1– 6. doi:10.1136/ tional socialization. Journal of Vocational Behavior, 42, 170 –183. doi:
qshc.2008.030262 10.1006/jvbe.1993.1012
Lipnack, J., & Stamps, J. (2000). Virtual teams: People working across Pearce, C. L. (2004). The future of leadership: Combining vertical and
boundaries with technology. New York, NY: Wiley. shared leadership to transform knowledge work. Academy of Manage-
Lowe, K. B., Kroeck, K. G., & Sivasubramaniam, N. (1996). Effectiveness ment Executive, 18, 47–57. doi:10.5465/AME.2004.12690298
correlates of transformational leadership and transactional leadership: A Pearce, C. L., & Conger, J. A. (2003). Shared leadership: Reframing the
meta-analytic review of the MLQ literature. The Leadership Quarterly, hows and whys of leadership. Thousand Oaks, CA: Sage.
Pearce, C. L., & Manz, C. C. (2005). The new silver bullets of leadership:
7, 385– 425. doi:10.1016/S1048-9843(96)90027-2
The importance of self and shared leadership in knowledge work.
MacCallum, R. C., Zhang, S., Preacher, K. J., & Rucker, D. D. (2002). On
Organizational Dynamics, 34, 130 –140. doi:10.1016/j.orgdyn.2005.03
the practice of dichotomization of quantitative variables. Psychological
.003
Methods, 7, 19 – 40. doi:10.1037/1082-989X.7.1.19
Pearce, C. L., & Sims, H. P., Jr. (2002). Vertical versus shared leadership
Majchrzak, A., Malhotra, A., Stamps, J., & Lipnack, J. (2004, May). Can
as predictors of the effectiveness of change management teams: An
absence make a team grow stronger? Harvard Business Review, 1–9.
examination of aversive, directive, transactional, transformational, and
Malhotra, A., & Majchrzak, A. (2004). Enabling knowledge creation in
empowering leader behaviors. Group Dynamics: Theory, Research, and
far-flung teams: Best practices for IT support and knowledge sharing. Practice, 6, 172–197. doi:10.1037/1089-2699.6.2.172
Journal of Knowledge Management, 8, 75– 88. doi:10.1108/ Pearce, C. L., Yoo, Y., & Alavi, M. (2004). Leadership, social work and
13673270410548496 virtual teams: The relative influence of vertical versus shared leadership
Malhotra, A., Majchrzak, A., & Rosen, B. (2007). Leading virtual teams. in the nonprofit sector. In R. E. Riggio, S. Smith-Orr, & J. Shakely
The Academy of Management Perspectives, 21, 60 –70. doi:10.5465/ (Eds.), Improving leadership in nonprofit organizations (pp. 180 –203).
AMP.2007.24286164 San Francisco, CA: Jossey-Bass.
Manz, C. C. (1986). Self-leadership: Toward an expanded theory of self- Peters, L., & Karren, R. J. (2009). An examination of the roles of trust and
influence processes in organizations. The Academy of Management functional diversity on virtual team performance ratings. Group &
Review, 11, 585– 600. Organization Management, 34, 479 –504. doi:10.1177/
Martins, L. L., Gilson, L. L., & Maynard, M. T. (2004). Virtual teams: 1059601107312170
What do we know and where do we go from here? Journal of Manage- Podsakoff, P. M., & MacKenzie, S. B. (1994). An examination of the
ment, 30, 805– 835. doi:10.1016/j.jm.2004.05.002 psychometric properties and nomological validity of some revised and
 LEADING VIRTUAL TEAMS 403

reduced “substitutes for leadership” scales. Journal of Applied Psychol- Staples, D., & Zhao, L. (2006). The effects of cultural diversity in virtual
ogy, 79, 702–713. doi:10.1037/0021-9010.79.5.702 teams versus face-to-face teams. Group Decision and Negotiation, 15,
Podsakoff, P. M., MacKenzie, S. B., & Bommer, W. H. (1996). Meta- 389 – 406. doi:10.1007/s10726-006-9042-x
analysis of the relationships between Kerr and Jermier’s substitutes for Townsend, A. M., DeMarie, S. M., & Hendrickson, A. R. (1998). Virtual
leadership and employee job attitudes, role perceptions, and perfor- teams: Technology and the workplace of the future. Academy of Man-
mance. Journal of Applied Psychology, 81, 380 –399. doi:10.1037/0021- agement Executive, 12, 17–29.
9010.81.4.380 Tsui, A. S., Nifadkar, S. S., & Ou, A. Y. (2007). Cross-national, cross-
Purvanova, R. K., & Bono, J. E. (2009). Transformational leadership in cultural organizational behavior research, advances, gaps, and recom-
context: Face-to-face and virtual teams. The Leadership Quarterly, 20, mendations. Journal of Management, 33, 426 – 478. doi:10.1177/
343–357. doi:10.1016/j.leaqua.2009.03.004 0149206307300818
Riley, S., & Wrench, D. (1985). Mentoring among women lawyers. Jour- Tyran, K. L., Tyran, C. G., & Shepherd, M. (2003). Exploring emerging
nal of Applied Social Psychology, 15, 374 –386. doi:10.1111/j.1559- leadership in virtual teams. In C. B. Gibson & S. G. Cohen (Eds.),
1816.1985.tb00913.x Virtual teams that work: Creating conditions for virtual team effective-
Ringle, C. M., Wende, S., & Will, A. (2006). SmartPLS (Version 2.0 M3). ness (pp. 183–195). San Francisco, CA: Jossey-Bass.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Retrieved from http://www.smartpls.de Van Der Vegt, G. S., Emans, B., & Van De Vliert, E. (2000). Team
Sambamurthy, V., & Chin, W. W. (1994). The effects of group attitudes members’ affective responses to patterns of intragroup interdependence
This document is copyrighted by the American Psychological Association or one of its allied publishers.

toward alternative GDSS designs on the decision-making performance and job complexity. Journal of Management, 26, 633– 655.
of computer-supported groups. Decision Sciences, 25, 215–241. doi: Wayne, S. J., Shore, L. M., & Liden, R. C. (1997). Perceived organiza-
10.1111/j.1540-5915.1994.tb01840.x tional support and leader–member exchange: A social exchange per-
Sarker, S., Anjuja, M., Sarker, S., & Kirkeby, S. (2011). The role of spective. Academy of Management Journal, 40, 82–111. doi:10.2307/
communication and trust in global virtual teams: A social network 257021
perspective. Journal of Management Information Systems, 28, 273–309. Whitely, W. T., Dougherty, T. W., & Dreher, G. F. (1991). Relationship of
doi:10.2753/MIS0742-1222280109 career mentoring and socioeconomic origin to managers’ and profes-
Scandura, T. A., & Ragins, B. R. (1993). The effects of sex and gender role sionals’ early career progress. Academy of Management Journal, 34,
orientation on mentorship in male-dominated occupations. Journal of 331–350. doi:10.2307/256445
Vocational Behavior, 43, 251–265. doi:10.1006/jvbe.1993.1046 Wiesenfeld, B. M., Raghuram, S., & Garud, R. (1999). Communication
Schminke, M., Cropanzano, R., & Rupp, D. E. (2002). Organization patterns as determinants of organizational identification in a virtual
structure and fairness perceptions: The moderating effects of organiza- organization. Organization Science, 10, 777–790. doi:10.1287/orsc.10.6
tional level. Organizational Behavior and Human Decision Processes, .777
89, 881–905. doi:10.1016/S0749-5978(02)00034-1 Wunderer, R. (2002). Fuehrung und Zusammenarbeit: Einer unternehmer-
Seers, A. (1989). Team-member exchange quality: A new construct for ische Fuehrungslehre [Leadership and collaboration: Organizational
role-making research. Organizational Behavior and Human Decision leadership instruction] (4th ed.). Neuwied, Germany: Kriftel.
Processes, 43, 118 –135. Yukl, G. (2010). Leadership in organizations (7th ed.). Englewood Cliffs,
Sherony, K. M., & Green, S. G. (2002). Coworker exchange: Relationships NJ: Prentice-Hall.
between coworkers, leader–member exchange, and work attitudes. Jour- Zaccaro, S. J., & Bader, P. (2003). E-leadership and the challenges of
nal of Applied Psychology, 87, 542–548. doi:10.1037/0021-9010.87.3 leading e-teams: Minimizing the bad and maximizing the good. Orga-
.542 nizational Dynamics, 31, 377–387. doi:10.1016/S0090-
Shin, Y. (2004). A person– environment fit model for virtual organizations. 2616(02)00129-8
Journal of Management, 30, 725–743. doi:10.1016/j.jm.2004.03.002 Zaccaro, S. J., Rittman, A. L., & Marks, M. A. (2001). Team leadership.
Siebdraht, F., Hoegl, M., & Ernst, H. (2009). How to manage virtual teams. The Leadership Quarterly, 12, 451– 483. doi:10.1016/S1048-
MIT Sloan Management Review, 50, 63– 69. 9843(01)00093-5
Sosik, J. J., Avolio, B. J., & Kahai, S. S. (1997). Effects of leadership style Zigurs, I. (2003). Leadership in virtual teams: Oxymoron or opportunity?
and anonymity on group potency and effectiveness in a group decision Organizational Dynamics, 31, 339 –351. doi:10.1016/S0090-
support system environment. Journal of Applied Psychology, 82, 89 – 2616(02)00132-8
103. doi:10.1037/0021-9010.82.1.89
Stanko, T. L., & Gibson, C. B. (2009). The role of cultural elements in
virtual teams. In R. S. Bhagat & R. M. Steers (Eds.), Cambridge Received August 27, 2011
handbook of culture, organization, and work (pp. 272–304). doi: Revision received July 9, 2012
10.1017/CBO9780511581151.012 Accepted July 18, 2012 䡲